This proposed R01 constitutes a collaboration among three ongoing longitudinal studies: the Seattle Social Development Project (SSDP, Karl G. Hill, PI), the Raising Healthy Children Project (RHC, Richard F. Catalano, PI) and projects from the Minnesota Center for Twin and Family Research (MCTFR, Matthew McGue, and William Iacono, PIs). To lay the foundation for the proposed study, work to be completed under the current two- year ARRA grant has permitted SSDP to (1) create measures of addiction phenotypes and measures of the environment suitable for gene-environment (G-E) analysis, (2) test models of environment and phenotype suitable for later inclusion of genetic data, (3) collect DNA samples from the SSDP longitudinal panel and (4) conduct a whole-genome genotyping in anticipation of subsequent genetic association and G-E analyses. The proposed R01 will extend this work by conducting genetic association analyses and examining G-E interplay in the development of addiction and HIV sexual risk behavior, tasks not supported by the ARRA grant. In Aim 1 we will utilize data from SSDP, RHC, MCTFR and three samples from the Genotypes and Phenotypes (dbGaP) archive to develop genetic system vulnerability scores for three systems related to addiction: dopamine, serotonin and GABA. Aim 1 also examines the contribution of these genetic system vulnerability scores to HIV sexual risk behavior, illicit drug use and crime. Aim 2 will examine mechanisms of genetic influence on addiction as well as gene-environment correlation and interaction through a test-and-replicate strategy employing SSDP, RHC and MCTFR samples. Aim 2 also examines the extent to which these gene- environment models predict HIV-sexual risk behavior, illicit drug use and crime. Aim 3 will examine gene x intervention interaction (GxI), and whether the influence of genetic factors on problem outcomes can be mitigated by social developmental preventive intervention during childhood. The proposed study builds on data sets of existing comparable phenotypic, individual and environmental measures in three well-characterized, longitudinal samples. Two of the data sets (SSDP and MCTFR) already have whole-genome genotypic data, and RHC has DNA collected and ready to be genotyped. All three panels contain indicators of environmental risk factors for tobacco and alcohol use and problem use assessed multiple times from childhood to adulthood. All three panels also include assessments of DSM-IV diagnoses as well as other quantitative measures of drug abuse and dependence and related phenotypes of demonstrated heritability. Finally, both SSDP and RHC implemented a social-developmental preventive intervention with parents, teachers and children early in life, permitting examination of gene x intervention interaction. The alliance of the SSDP, RHC and MCTFR research teams provides an opportunity to learn how, at different developmental periods, environmental factors might amplify or reduce genetic vulnerability to tobacco and alcohol problems. Findings will inform the design of improved preventive and treatment interventions for these common and costly disorders. 2.